python dataclass与dataclass_json修饰器的理解

实例：

from dataclasses import dataclass
from dataclasses_json import dataclass_json
from typing import Any, Dict, List, Tuple

@dataclass_json
@dataclass
class Nn:
    xycoords: str

@dataclass_json
@dataclass
class Station:
    stations: Tuple[Nn, ...]

@dataclass_json
@dataclass
class Busline_list:
    busline_list: Tuple[Station, ...]

@dataclass_json
@dataclass
class Data:
    data: Tuple[Busline_list, ...]

with open("test.json", "r", encoding="utf-8") as fp:
    a = Data.from_json(fp.read())

结果：
Data(data=(Busline_list(busline_list=(Station(stations=(Nn(xycoords=‘1’), Nn(xycoords=‘2’))), Station(stations=(Nn(xycoords=‘3’), Nn(xycoords=‘4’))))), Busline_list(busline_list=(Station(stations=(Nn(xycoords=‘5’), Nn(xycoords=‘6’))), Station(stations=(Nn(xycoords=‘7’), Nn(xycoords=‘8’)))))))

---

可以发现递归调用了所有定义的类，将json文件全部解析为定义的类型

---

假如将上述代码修改为以下：

from dataclasses import dataclass
from dataclasses_json import dataclass_json
from typing import Any, Dict, List, Tuple

@dataclass_json
@dataclass
class Nn:
    xycoords: str

@dataclass_json
@dataclass
class Station:
    stations: Tuple[Nn, ...]

@dataclass_json
@dataclass
class Busline_list:
    busline_list: Tuple[int, ...]  # 将Station改为int

@dataclass_json
@dataclass
class Data:
    data: Tuple[Busline_list, ...]

with open("test.json", "r", encoding="utf-8") as fp:
    a = Data.from_json(fp.read())

结果为:
Data(data=(Busline_list(busline_list=({‘stations’: [{‘xycoords’: ‘1’}, {‘xycoords’: ‘2’}]}, {‘stations’: [{‘xycoords’: ‘3’}, {‘xycoords’: ‘4’}]})), Busline_list(busline_list=({‘stations’: [{‘xycoords’: ‘5’}, {‘xycoords’: ‘6’}]}, {‘stations’: [{‘xycoords’: ‘7’}, {‘xycoords’: ‘8’}]}))))

---

发现最终只递归到busline_list这一类，没有对json进行进一步深层次的解析